© 2015. Published by The Company of Biologists Ltd | Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

RESEARCH ARTICLE STEM CELLS AND REGENERATION Rab8a vesicles regulate Wnt ligand delivery and Paneth cell maturation at the intestinal stem cell niche Soumyashree Das1, Shiyan Yu1, Ryotaro Sakamori1, Pavan Vedula1, Qiang Feng1, Juan Flores1, Andrew Hoffman2, Jiang Fu3, Ewa Stypulkowski1, Alexis Rodriguez1, Radek Dobrowolski1, Akihiro Harada4, Wei Hsu3, Edward M. Bonder1, Michael P. Verzi2,5 and Nan Gao1,5,*

ABSTRACT the cytoplasmic protein Dishevelled (Dvl, or Dsh) (Wong et al., Communication between stem and niche supporting cells maintains 2003), in association with low-density lipoprotein receptor-related the homeostasis of adult tissues. Wnt signaling is a crucial regulator protein 5 and 6 (Lrp5/6), triggering the assembly of a multiprotein of the stem cell niche, but the mechanism that governs Wnt complex at the plasma membrane (Bilic et al., 2007). This plasma membrane-localized protein aggregate, which is sometimes referred ligand delivery in this compartment has not been fully investigated. ‘ ’ We identified that Wnt secretion is partly dependent on Rab8a- to as the Wnt signalosome (Bilic et al., 2007), inactivates a mediated anterograde transport of Gpr177 (wntless), a Wnt-specific cytoplasmic destruction machinery consisting of casein kinase 1, transmembrane transporter. Gpr177 binds to Rab8a, depletion of glycogen synthase kinase 3 (Gsk3), axis inhibitor (Axin), adenomatosis polyposis coli and the E3 ubiquitin ligase β-Trcp which compromises Gpr177 traffic, thereby weakening the secretion β of multiple Wnts. Analyses of generic Wnt/β-catenin targets in Rab8a (Btrc), causing -catenin stabilization (Cadigan and Peifer, 2009; knockout mouse intestinal crypts indicate reduced signaling activities; Huang and He, 2008; MacDonald et al., 2009) and transcriptional maturation of Paneth cells – a Wnt-dependent cell type – is severely activation of Wnt targets (He et al., 2004; MacDonald and He, 2012; affected. Rab8a knockout crypts show an expansion of Lgr5+ and Tamai et al., 2000; Wehrli et al., 2000). This signaling cascade, which Hopx+ cells in vivo. However, in vitro, the knockout enteroids exhibit is often referred to as the canonical Wnt pathway, plays a fundamental significantly weakened growth that can be partly restored by role in fetal development and adult tissue homeostasis (Clevers, exogenous Wnts or Gsk3β inhibitors. Immunogold labeling and 2006; Clevers and Nusse, 2012). Inappropriate activation of this surface protein isolation identified decreased plasma membrane pathway in diseases, especially colon cancers, has highlighted its localization of Gpr177 in Rab8a knockout Paneth cells and profound influence on cellular behavior (Angers and Moon, 2009; fibroblasts. Upon stimulation by exogenous Wnts, Rab8a-deficient Clevers and Nusse, 2012; de Lau et al., 2007; MacDonald et al., 2009; cells show ligand-induced Lrp6 phosphorylation and transcriptional Nusse et al., 2008; Polakis, 2007; Reya and Clevers, 2005). Certain reporter activation. Rab8a thus controls Wnt delivery in producing Wnt-Fzd complexes activate non-canonical Wnt pathways and cells and is crucial for Paneth cell maturation. Our data highlight the regulate cell migration and polarity via Rho subfamily small profound tissue plasticity that occurs in response to stress induced by GTPases (Boutros and Mlodzik, 1999; Eaton et al., 1996; Fanto depletion of a stem cell niche signal. et al., 2000; Habas et al., 2003, 2001; Sakamori et al., 2014; Strutt et al., 1997; Wallingford and Habas, 2005). KEY WORDS: Rab8a, Gpr177, Wntless, Wnt secretion, Intestinal In Wnt-producing cells, newly synthesized Wnt proteins are lipid stem cell, Paneth cell modified in the endoplasmic reticulum (ER) by an acyltransferase, Porcupine (Takada et al., 2006), and transported by the multi-pass INTRODUCTION transmembrane protein Wntless [also known as G protein-coupled Wnts are cysteine-rich glycolipoproteins that act as paracrine or receptor 177 (Gpr177) in mammals] for exocytosis (Bänziger et al., autocrine ligands believed to engage in short-range signaling (Willert 2006; Bartscherer et al., 2006). Global or tissue-specific ablation of et al., 2003; Willert and Nusse, 2012). Signal transduction in the Wnt- Wntless/Gpr177 in various animals causes phenotypes that responding cell is initiated by binding of Wnts to their seven-pass resemble loss of Wnts (Bänziger et al., 2006; Bartscherer et al., transmembrane Frizzled (Fzd) receptors (Schulte, 2010; Schulte and 2006; Fu et al., 2011), leading to the current notion that Wntless/ Bryja, 2007; Wu and Nusse, 2002). The Wnt-Fzd complex recruits Gpr177 represents the specific and possibly sole transporter for secretion of most Wnts (Ching and Nusse, 2006; Das et al., 2012; Port and Basler, 2010). After Wnt release at the cell surface, Gpr177 1Department of Biological Sciences, Rutgers University, Newark, NJ 07102, USA. is internalized from the plasma membrane via a Clathrin-dependent 2Department of Genetics, Human Genetics Institute of New Jersey, Rutgers University, Piscataway, NJ 08854, USA. 3Department of Biomedical Genetics, pathway to endosomes (Gasnereau et al., 2011; Pan et al., 2008), Center for Oral Biology, James P. Wilmot Cancer Center, Stem Cell and where it is retrieved in a retromer Vps35- and Snx3-dependent Regenerative Medicine Institute, University of Rochester Medical Center, fashion to the Golgi for new rounds of Wnt transport (Belenkaya Rochester, NY 14642, USA. 4Department of Cell Biology, Graduate School of Medicine, Osaka University 2-2, Yamadaoka, Suita, Osaka 565-0871, Japan. et al., 2008; Franch-Marro et al., 2008; Harterink et al., 2011; Port 5Rutgers Cancer Institute of New Jersey, New Brunswick, NJ 08901, USA. et al., 2008; Rojas et al., 2008; Yang et al., 2008). In contrast to the intensive study of retrograde traffic of Gpr177, the cellular *Author for correspondence ([email protected]) machineries that govern anterograde Gpr177-Wnt transport have This is an Open Access article distributed under the terms of the Creative Commons Attribution not been fully explored. It is unclear whether Gpr177-Wnt follows a License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed. vesicular membrane bulk flow (Hong and Tang, 1993) or if Wnt exocytosis is subject to regulation by specific secretory machinery

Received 16 December 2014; Accepted 16 April 2015 (Das et al., 2012). DEVELOPMENT

2147 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

Among various cell types that produce distinct cohorts of Wnts et al., 2007). However, the contribution of Rab8 vesicles to surrounding the intestinal stem cell niche (Gregorieff et al., 2005), intestinal crypt homeostasis is not defined. Paneth cells are the major epithelial Wnt producers, expressing A recent screening for Rab modulators of the Wnt pathway Wnt3, Wnt6 and Wnt9b, and co-occupy the crypt bottom with stem identified RAB8B, but not RAB8A, as a crucial regulator of cells. The self-renewal of fast-cycling Lgr5+ stem cells (Barker canonical Wnt signaling in receiving cells by directly interacting et al., 2007), as shown by organoid-forming capacity in culture, is with LRP6 and CK1γ (Demir et al., 2013). We provide evidence enhanced by close association with Paneth cells or by addition of here that, in Wnt-producing cells, Rab8a regulates Gpr177 exogenous Wnt ligands (Sato et al., 2011). The ablation of Paneth anterograde traffic and Wnt secretion. Using immunogold cells in several mouse models caused recoverable loss of Lgr5+ stem labeling of endogenous Gpr177 in native Wnt producers, Wnt cells (Sato et al., 2011). However, Atoh1-deficient mouse intestines secretion and reporter assays, we demonstrate that Rab8a ablation with an absence of Paneth cell differentiation preserved functional impairs Gpr177 trafficking in Wnt producers, attenuating Wnt intestinal epithelia (Durand et al., 2012; Kim et al., 2012), hinting at secretion and canonical Wnt signaling in vivo and ex vivo. Rab8a a high plasticity of crypt cells (Tetteh et al., 2014). In addition to knockout intestinal crypts showed altered cell organization in Paneth cells, subepithelial stromal cells express Wnt2b, Wnt4 and response to decreased extracellular Wnts in the niche. These data Wnt5a (Farin et al., 2012; Gregorieff et al., 2005; Miyoshi et al., shed light on intestinal crypt plasticity in response to stress induced 2012). Wnt5a+ mesothelial cells contribute to regenerating nascent by defective niche signal traffic. crypts after tissue injury (Miyoshi et al., 2012). Intestinal epithelia- specific ablation of Wnt3 (Farin et al., 2012), or Porcupine deletion RESULTS in both epithelia and myofibroblasts (Kabiri et al., 2014; San Roman Gpr177 traffics through Rab8a vesicles et al., 2014), did not cause detectable tissue damage. In culture, We established a stable Henrietta Lacks (HeLa) human cell line Wnt3-deficient intestinal organoids fail to propagate, but expressing 3×Flag-GPR177 to identify regulators for Wnt-GPR177 administration of Wnts was able to restore the growth (Farin et al., trafficking. Using cell lysates extracted in the presence of 1% Triton 2012), collectively suggesting that multiple sources of Wnts X-100, we performed co-immunoprecipitation analyses to identify redundantly support the stem cell niche. potential interactions between GPR177 and key trafficking Upon quantitative loss of Lgr5+ stem cells, a proposed ‘reserve’ regulators. We detected association of GPR177 with RAB5, stem cell pool rejuvenates the epithelia (Tian et al., 2011). These RAB8A and RAB9 (Fig. 1A). As GPR177 is internalized into stem cells are considered to be slow-cycling and are identified by endosomes (Belenkaya et al., 2008) during retrograde trafficking, several markers, including Bmi1 (Sangiorgi and Capecchi, 2008), association of GPR177 with RAB5 and RAB9 reflected endocytosis Hopx (Takeda et al., 2011), Lrig1 (Powell et al., 2012) and Tert of GPR177 (Gasnereau et al., 2011). Association between GPR177 (Montgomery et al., 2011). Lineage conversion from these and the RAB8A vesicular compartment has not been described. proposed quiescent cells to Lgr5+ cells has been observed during Given that RAB8 transports several G protein-coupled receptors homeostasis or epithelial injury (Takeda et al., 2011; Tian et al., (GPCRs) (Dong et al., 2010; Esseltine et al., 2012), we postulated 2011; Yan et al., 2012). However, all these markers are highly that RAB8A vesicles might be involved in anterograde traffic of the expressed in Lgr5+ cells (Muñoz et al., 2012; Wong et al., 2012) and Wnt-GPR177 complex. Of note, under similar conditions, 3×Flag- also in a subset of Lgr5+ label-retaining cells (LRCs) (Buczacki GPR177 was not detected in association with RAB7, RAB11 or et al., 2013). This subset of Lgr5+ cells was recently proposed to VPS35 (Fig. 1A), suggesting that GPR177 and RAB8A might exist constitute secretory precursors for Paneth and enteroendocrine cells in a relatively stable detergent-resistant complex. The interaction and could be reactivated by injury for epithelial regeneration between GPR177 and RAB8A was likely to be physiologically (Buczacki et al., 2013; Roth et al., 2012). In a parallel study, Dll1+ relevant as a truncated GPR177 lacking the C-terminal cytoplasmic secretory precursors were shown to revert to stem cells upon injury tail (GPR177Δ44) failed to associate with RAB8A (Fig. 1B). Using to regenerate the epithelia (van Es et al., 2012). These secretory glutathione S-transferase (GST)-RAB8A fusion proteins, we precursors are postulated to represent the proposed ‘+4’ quiescent performed GST pull-down assays using 3×Flag-GPR177 cell cells (Tetteh et al., 2014). However, in contrast to Lgr5+ cells, lysates and consistently detected binding of GPR177 to GST- Bmi1+ cells were shown to resist Wnt perturbation and radiation RAB8A but not to GST, GST-CDC42 or GST-synaptotagmin-like 1 injury (Yan et al., 2012). Thus, whether a dedicated quiescent stem (JFC)-D1 (Fig. 1C), suggesting that RAB8A and GPR177 indeed cell population truly exists is still under debate (Tetteh et al., 2014). associate in a complex. When GPR177-mCherry and EGFP- Vesicular trafficking influences Wnt signaling capacities in both RAB8A were transiently expressed in HeLa cells (Fig. 1D) or ligand-producing and ligand-receiving cells (de Groot et al., 2013; human colonic epithelial Caco2 cells (supplementary material Fig. Feng and Gao, 2014; Kabiri et al., 2014; San Roman et al., 2014). S1A), three populations of vesicles – mCherry positive, EGFP During intestinal differentiation, the intestinal cell fate activator positive and mCherry/EGFP double positive – were observed and Cdx2 transcriptionally regulates the expression of Rab8 small confirmed by line scans, indicating that some GPR177 traffics GTPases, which are members of the Ras superfamily (Gao and through RAB8A vesicles (Fig. 1D). Kaestner, 2010). Rab8 directly binds isoforms of the myosin V Rab8a is required for docking vesicular cargo on the myosin V motor (Hume et al., 2001; Rodriguez and Cheney, 2002; Roland motor for exocytotic transport (Khandelwal et al., 2013; Roland et al., 2009), facilitating exocytotic cargo movements on actin tracks et al., 2011). We derived Rab8a−/− and wild-type mouse embryonic in epithelial and non-epithelial cells (Ang et al., 2003; Bryant et al., fibroblasts (MEFs) and transiently expressed GPR177-mCherry in 2010; Gerges et al., 2004; Hattula et al., 2006; Henry and Sheff, these cells to track dynamic vesicle movement. Rab8a−/− MEFs 2008; Huber et al., 1993a,b; Sato et al., 2009; Sun et al., 2014). showed no qualitative difference in terms of peri-nuclear localization Global Rab8a ablation in mice impairs the apical delivery of of Gpr177+ vesicles when compared with wild-type MEFs of similar peptidases and nutrient transporters to enterocyte brush borders; as a cell morphology (Fig. 1E; supplementary material Fig. S1B). consequence, these proteins are transported into lysosomes, causing However, by binning instantaneous speeds of Gpr177 vesicles in nutrient deprivation and postnatal death of knockout mice (Sato 70 nm/s increments and analyzing the frequency distribution as a DEVELOPMENT

2148 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

Fig. 1. RAB8A intersects GPR177 traffic. (A) Flag-GPR177 was immunoprecipitated (IP) from lysates of a stable human HeLa cell line in the presence of 1% Triton X-100. Precipitates were blotted (IB) for various vesicular markers. (B) Flag-GPR177Δ44 lacking the C-terminal tail failed to co-immunoprecipitate with RAB8A. (C) GST pull-down showed binding of Flag-GPR177 to GST-RAB8A, but not to GST, GST-CDC42 or GST-JFC-D1. Data are representative of three independent experiments. (D) Live cell imaging detected GPR177-mCherry in EGFP-RAB8Avesicles in HeLa cells. The line scan histogram (along the dotted line − − in the merge) shows colocalization of the two fluorescent signals (arrows). (E) Single vesicle tracks of GPR177-mCherry vesicles in Rab8a+/+ (top) and Rab8a / (bottom) MEFs (t=10 s; n=8 vesicles in each cell). Arrows indicate the start and end of an individual vesicle track. See supplementary material Fig. S1B, Movies 1 and 2. (F) Instantaneous vesicle speed distribution (nm/s), with a bin size of 70 nm/s. The speed of Gpr177 vesicles in Rab8a+/+ cells peaks at 130-140 nm/s − − (green dashed line), agreeing with myosin V-powered movement (bar), which was reduced in Rab8a / MEFs (red dashed curve). 1717 steps for wild type and − − 995 steps for Rab8a / were analyzed. Scale bars: 10 μm. percentage of total steps (n=1717 for wild type and n=995 for Impaired Wnt secretion in Rab8a−/− cells Rab8a−/−; Fig. 1F; supplementary material Movies 1 and 2), Impaired Gpr177 traffic might influence its function in Wnt we observed a reduction in the frequency of instantaneous speeds exocytosis. As Wnt5a is endogenously expressed by MEFs (Sato within 106-176 nm/s (centered at 140 nm/s) in Rab8a−/− MEFs. et al., 2004) and is known to influence non-canonical and canonical Myosin V-powered melanosome transport has been reported to have Wnt pathways in mammalian cells (He et al., 1997; Mikels and an average instantaneous speed of 140 nm/s (Wu et al., 1998). Curve Nusse, 2006; Okamoto et al., 2014), we compared Wnt5a/b fitting that identified peaks of frequency distributions showed that secretion in Rab8a+/+ and Rab8a−/− MEFs. Using a Wnt5a/b- the instantaneous speed of Gpr177+ vesicles in Rab8a−/− cells specific antibody, we detected significant reductions in secreted decreased to 53 nm/s, as compared with 105 nm/s in Rab8a+/+ cells Wnt5a/b in media conditioned by Rab8a−/− MEFs (Fig. 2A). (dashed lines, Fig. 1F). An increased number of Gpr177+ vesicles in Rab8a−/− MEFs accumulated more intracellular Wnt5a/b, as the slow (1-50 nm/s) and fast (250-500 nm/s) speed ranges in indicated in cell lysates (Fig. 2A), suggesting that a blockage of Rab8a−/− cells suggested diffusion-like movement and microtubule- Wnt5a/b secretion might have caused ligand accumulation. When dependent transport, respectively (Wu et al., 1998). These results exogenous WNT5A was transiently transfected into Rab8a+/+ suggest that the absence of Rab8a vesicles affected Gpr177 MEFs, it enhanced Wnt Topflash reporter activity (Fig. 2B), trafficking dynamics, which was possibly attributable to the suggesting elevated WNT5A autocrine signaling (Goel et al., 2012). reduced myosin V-powered vesicle movement (Roland et al., However, similar overexpression of WNT5A in Rab8a−/− MEFs

2011; Sun et al., 2014). failed to augment Wnt reporter activity (Fig. 2B), suggesting that DEVELOPMENT

2149 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

− − Fig. 2. Rab8a deletion reduces Wnt secretion. (A) Concentrated conditioned media or cell lysates, in increasing amounts, from Rab8a+/+ and Rab8a / MEFs − − were analyzed by western blot. Rab8a / MEFs showed less secreted, but more intracellular, Wnt5a/b. The ratio of secreted to intracellular Wnt5a/b was deduced from corresponding samples (n=3, bar chart). Histone H3 was used to detect medium contamination by cell lysates and to normalize corresponding Wnt5a/b bands. (B) WNT5A expression constructs or empty vectors were transiently transfected together with Topflash reporter and Renilla luciferase plasmids into − − − − Rab8a+/+ and Rab8a / MEFs, followed by dual-luciferase assays. Rab8a+/+ and Rab8a / MEFs showed 14.6-fold and 3.9-fold inductions of Topflash activities, − − respectively, by transfected WNT5A as compared with empty vector-transfected counterparts. **P<0.01. (C) Rab8a / MEFs secrete less Wnt3a-Gluc. Wnt3a- − − − − − − Gluc was transiently transfected into wild-type, Rab8a / , Rab8b knockdown (KD), Rab8a / /Rab8b KD and Gpr177 / MEFs, with firefly luciferase serving as control for transfection efficiency. Secretion of Wnt3a-Gluc was eliminated by 5 mM C59 or Gpr177 depletion. **P<0.01, ***P<0.001; n.s., not significant. − − (D) Rab8a / MEFs showed insignificant changes in the secretion of Shh-Renilla, Met-Luc or the biosynthetic cargo alkaline phosphatase (AP). (E) Surface − − − − protein biotinylation and isolation detected decreased Gpr177 in Rab8a / MEFs. Consistent results were obtained in two independent experiments. (F) Rab8a / MEFs showed similar levels of cell surface Fzd (1-10) and Lrp6 compared with Rab8a+/+ MEFs. Note that Wnt3a stimulated surface Lrp6 phosphorylation − − − − (Ser1490) in Rab8a+/+ and Rab8a / MEFs. (G) Topflash assays showed that Rab8a / MEFs, with a significantly lower basal Wnt signaling activity (##P<0.01, compared with vehicle-treated Rab8a+/+ MEFs), responded strongly to Wnt3a stimulation. **P<0.01, ***P<0.001, compared with vehicle-treated cells.

Rab8a−/− MEFs might not properly secrete the transfected compared with Rab8a+/+ MEF-conditioned media (Fig. 2C). No WNT5A. reduction of Wnt3a-Gluc secretion was detected for MEFs with We examined whether Rab8a deficiency also impaired secretion stable Rab8b depletion by a lentiviral shRNA against Rab8b of canonical Wnt ligands such as Wnt3a. MEFs do not detectably (Fig. 2C). Combined depletion of Rab8a and Rab8b did not elicit express endogenous Wnt3a. Wnt3a-Gluc, a fusion protein additive inhibition of Wnt3a-Gluc secretion as compared with comprising Wnt3a and Gaussia luciferase (Chen et al., 2009), has Rab8a deletion alone (Fig. 2C), suggesting that Rab8b, neither by been shown to act as a functional ligand and has been successfully itself nor in combination with Rab8a, influences Wnt3a-Gluc used to screen for small molecular inhibitors of Porcupine (Chen secretion. This is in agreement with the primary function of RAB8B et al., 2009). We transiently transfected Rab8a+/+ and Rab8a−/− in ligand-receiving cells (Demir et al., 2013). Stronger inhibitory MEFs with Wnt3a-Gluc and assessed the conditioned media by effects on Wnt3a-Gluc secretion were observed upon treatment with luciferase assay (Chen et al., 2009). We detected an ∼80% reduction C59 (∼93% reduction), a Porcupine inhibitor (Chen et al., 2009), or −/− of secreted Wnt3a-Gluc in Rab8a MEF-conditioned media as by Gpr177 ablation in MEFs (∼97% reduction) (Fig. 2C). Notably, DEVELOPMENT

2150 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046 in contrast to the near-complete loss of Wnt3a-Gluc secretion majority of surviving Rab8a−/− organoids showed tiny buds and large caused by C59 treatment or Gpr177 deletion, 19% of Wnt3a-Gluc lumens (Fig. 3E and Fig. 4A), similar to Wnt3- or Atoh1-deficient proteins were still detected in Rab8a−/− MEF-conditioned media enteroids (Durand et al., 2012; Farin et al., 2012). Administration of the (Fig. 2C), suggesting that Rab8a deletion partially compromised Porcupine inhibitor C59 to surviving Rab8a−/− organoids further Wnt secretion, observations that are consistent with the data on abolished epithelial budding, followed by the near-complete Wnt5a/b secretion (Fig. 2A). We further examined the secretory disappearance of existing buds within 4 days of treatment (101 out activities of several non-Wnt ligands, including sonic hedgehog of 105 Rab8a-deficient organoids; supplementary material Fig. S2A). (Shh)-Renilla luciferase (Ma et al., 2002) as another important A fraction of C59-treated organoids maintained cyst-like structures; morphogen, the biosynthetic cargo secreted alkaline phosphatase however, their proliferative activities were markedly reduced as (Yu et al., 2014c) and the constitutively secreted cargo Metridia assessedby5-ethynyl-2′-deoxyuridine (EdU) labeling (supplementary luciferase (Markova et al., 2004). None of these showed significant material Fig. S2B). C59 treatment presumably depleted residual Wnt secretory abnormalities in Rab8a−/− MEFs (Fig. 2D), suggesting a secretion from Rab8a-deficient organoids, suggesting that Rab8a degree of cargo selectivity by Rab8a. ablation partially, but not completely, blocked Wnt secretion. Importantly, when we supplemented culture media with Rab8a deletion reduces anterograde transport of Gpr177 exogenous Wnt3a, the survival and propagation capacities of Rab8a vesicles control the apical transport of brush border enzymes Rab8a-deficient organoids were significantly improved by ∼50% in enterocytes (Sato et al., 2014, 2007) and are responsible for (Fig. 4B,D). Wnt3a administration induced Rab8a+/+ organoids transporting cargoes for exocytosis (Khandelwal et al., 2013; Sato into a cyst-like morphology (Yin et al., 2014), an effect that was et al., 2007; Sun et al., 2014). To test whether Rab8a is necessary also observed for Rab8a-deficient organoids. Treating Rab8a- for anterograde transport of Gpr177 to the cell surface, we directly deficient organoids with CHIR99021, a Gsk3β inhibitor, improved measured surface-localized Gpr177 by surface-protein biotinylation survival by 46% (Fig. 4C,D), supporting the notion that the and isolation. We detected an ∼46% reduction of surface Gpr177 poor growth of Rab8a−/− organoids was due to insufficient Wnts in Rab8a−/− MEFs (Fig. 2E). In contrast to Gpr177, the amount of and that Rab8a-deficient cells can properly transduce the Wnt surface-localized Fzd receptors and Lrp6 co-receptor did not signal. change (Fig. 2F). Most importantly, addition of Wnt3a proteins to − − −/− cultured Rab8a / MEFs markedly stimulated cell surface Lrp6 Paneth cell maturation is severely impaired in Rab8a mice phosphorylation (Ser1490) to a level equivalent to that of Rab8a+/+ Paneth cells are the major epithelial Wnt producers within the cells (Fig. 2F), suggesting that Rab8a−/− cells can properly respond intestinal crypts (Sato et al., 2011). Lysozyme staining identified to exogenous ligand stimulation by assembling an Lrp6-containing significant reduction of lysozyme+ Paneth cells in Rab8a−/− crypts. surface protein complex (Bilic et al., 2007). In Topflash reporter Eighty-five percent of Rab8a−/− intestinal crypts contained virtually assays, Rab8a−/− MEFs showed lower basal reporter activities in no detectable lysozyme+ Paneth cells, while the rest had a single serum-deprived conditions (white bars in Fig. 2G), and again lysozyme+ cell (Fig. 5A). The decrease in mature lysozyme+ Paneth responded strongly to exogenous Wnt3a (Fig. 2G). cells re-established the notion that the maturation of this cell type depends on proper Wnt signaling (van Es et al., 2005). Interestingly, − − Reduced Wnt/β-catenin signaling in Rab8a−/− intestinal crypts when compared with Rab8a+/+, Rab8a / intestines transcribed Intestinal crypt homeostasis relies on proper Wnt signaling in the similar or even higher levels of several Paneth cell-specific genes, stem cell niche (Clarke, 2006; Gregorieff and Clevers, 2005; namely Lyz1, Mmp7 and Defa5 (Fig. 5B), suggesting that Rab8a Haegebarth and Clevers, 2009). Quantitative RT-PCR for canonical deletion might have actually blocked the terminal differentiation of Wnt targets showed decreased Axin2 and Ascl2 expression in Paneth cells from the precursors. Such intermediate secretory Rab8a−/− intestines (Fig. 3A). Western blots detected reduced precursors have recently been described (Buczacki et al., 2013; levels of total β-catenin, Tcf1 (T-cell factor 1; also known as Tcf7), Clevers et al., 2014). In cultured organoids, lysozyme staining Tcf4 and Sox9 (Fig. 3B). However, the level of c-Myc, a marker of identified Paneth cells in small buds of surviving Rab8a−/− transit amplifying cells (Gregorieff et al., 2005), increased in organoids (green, Fig. 5C), hinting that the preservation of a Rab8a−/− intestines (Fig. 3B). Immunohistochemical analyses of β- small number of Paneth cells in surviving knockout crypts might catenin showed a reduced number of crypts with nuclear β-catenin+ have facilitated their survival, whereas crypts totally lacking mature cells (only Paneth cell-containing crypts were compared, Fig. 3C). Paneth cells were arrested (red line, Fig. 4A). Within a single Rab8a−/− crypt, the number of nuclear β-catenin+ Analyses of Rab8a−/− intestinal crypts by transmission electron cells was also reduced compared with wild types, collectively microscopy (TEM) revealed substantial subcellular defects in these suggesting reduced Wnt/β-catenin signaling in knockout crypts. residual Paneth cells. Virtually all Rab8a−/− Paneth cells contained We then used the Axin2lacZ/+ reporter allele to determine fewer electron-dense secretory granules than typically found in this canonical Wnt signaling activity (Lustig et al., 2002). By cell type (Fig. 5D). Rab8a−/− Paneth cells also showed markedly establishing Rab8a−/−;Axin2lacZ/+ mice, we observed reduced expanded (∼2.5-fold) smooth ER cisternal stacks as compared with Axin2lacZ/+ reporter activities, as indicated by β-gal staining of those in Rab8a+/+ Paneth cells (Fig. 5E; supplementary material mouse small intestines, compared with Rab8a+/+;Axin2lacZ/+ Figs S3 and S4). littermates (Fig. 3D). Various non-Paneth cells surrounding crypts produce Wnts, and We further analyzed Axin2lacZ/+ reporter activity in cultured thus redundant sources of Wnts have been proposed to support the intestinal organoids. Rab8a−/−;Axin2lacZ/+ organoids also showed stem cell niche (Farin et al., 2012; Gregorieff et al., 2005). Rab8a reduced β-gal activities in epithelial buds when compared with those in global knockout (Rab8a−/−) presumably affected Wnt secretion Rab8a+/+;Axin2lacZ/+ epithelial buds (Fig. 3E). Remarkably, Rab8a−/− from all niche supporting cells. We postulated that intestinal intestinal organoids, as well as Rab8aloxP/loxP;Villin-Cre (Rab8aΔIEC) epithelial cell-specific Rab8a deletion (Rab8aΔIEC) would produce organoids, showed severely compromised growth and budding a milder phenotype than that of Rab8a−/− intestines. Indeed, −/− ΔIEC capability, with nearly 80% arrested within 2 days (Fig. 4A). The compared with Rab8a intestines, Rab8a intestines exhibited DEVELOPMENT

2151 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

Fig. 3. Rab8a deletion impairs canonical Wnt signaling in intestines. (A) Quantitative RT-PCR showed reduced Tcf1, Olfm4, Axin2 and Ascl2 expression in − − − − Rab8a / intestines. (B) Western blots showed reduced Tcf1, Tcf4, Sox9 and β-catenin levels in Rab8a / intestines. (C) Immunohistochemistry for β-catenin showed a reduced number of crypts with nuclear β-catenin+ cells (arrows). Note that only crypts with detectable Paneth cells were scored for their positive or negative inclusion of nuclear β-catenin (n=50 for each genotype). (D) β-Gal staining of mouse small intestines showed a significant reduction of Axin2 reporter − − activity in crypts in Axin2lacZ/+;Rab8a / mice. Thirty continuous crypts were analyzed in each section of independent wild-type and knockout mice. (E) β-Gal- − − stained Axin2lacZ/+ and Axin2lacZ/+;Rab8a / intestinal organoids showed significantly reduced bud number and size in the absence of Rab8a. Images were taken at day 10 after crypt plating. Arrow points to a small bud. Twenty organoids of each genotype were quantified for β-gal-stained bud areas (circled in red). *P<0.05, **P<0.01, ***P<0.001. Scale bars: 10 μm in C,D; 15 μminE. less pronounced Paneth cell loss. Continuous stretches of long-term survival compared with Rab8a−/− organoids (compare Rab8aΔIEC crypts containing Paneth cells existed next to crypts blue and red lines, Fig. 4A). without Paneth cells, suggesting an overall milder impact on Paneth cell maturation (supplementary material Fig. S5A,B). This Rab8a deletion decreases Gpr177 trafficking to the cell observation was echoed in organoid culture experiments. We surface in vivo observed two morphologically distinct clones of Rab8aΔIEC To further explore whether Gpr177 traffic was affected in Wnt organoids: one resembled wild type (supplementary material producers in vivo, we performed immunogold labeling of Gpr177 in Fig. S5C) with well-developed buds, whereas the other mimicked Rab8a+/+ and Rab8a−/− mouse intestines. The specificity of the Rab8a−/− organoids (Fig. 4A). The visibly improved epithelial Gpr177 antibody (Fu et al., 2009) was affirmed by a 90%, 91% budding in a fraction of Rab8aΔIEC organoids agreed with the and 93% reduction in gold particles in the ER, the non-ER observation that some Rab8aΔIEC crypts possessed more Paneth vesicular compartment and the plasma membrane, respectively, of ΔIEC cells than others. Nevertheless, Rab8a did not show improved Gpr177-deficient cells (supplementary material Fig. S6) as DEVELOPMENT

2152 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

Fig. 4. Rab8a deficiency-impaired organoid growth is restored by Wnt3a. (A) Crypts (n=100) of each genotype were seeded in triplicate and the number of − − Δ surviving organoids then counted daily. Growth of 80% of Rab8a / intestinal organoids was arrested within the first 2 days. Rab8a IEC did not show any − − significant improvement in organoid survival. Data were collected from three independent experiments. (B,C) After seeding, Rab8+/+ and Rab8a / intestinal organoids (n=100 for each genotype) were immediately supplemented with Wnt3a or the Gsk3β inhibitor CHIR at the indicated concentrations. Media containing − − these supplements were replenished daily. Rab8a / intestinal organoids showed similar morphological and proliferative features as Rab8+/+ organoids upon Wnt3a or CHIR treatment. (D) In the above experiments, surviving organoids were counted daily during 1 week after seeding. Percentages of surviving organoids − − are plotted. Note that Wnt3a and CHIR significantly increased the number of surviving Rab8a / organoids. Scale bars: 15 μm. compared with wild-type cells. Using identical labeling conditions, subepithelial stromal cells (supplementary material Fig. S9), which we detected an 11% total reduction of Gpr177+ gold particles in agreed with data from the biochemical isolation of surface Gpr177 Rab8a−/− Paneth cells. From TEM montage images we performed (Fig. 2E). quantitative analyses of gold particle distributions at the ER, plasma Given that Gpr177 is able to exit the ER, we postulated that membrane and non-ER Golgi/secretory vesicular compartments Rab8a absence might lead to mis-sorting of Gpr177 into (independent duodenal and jejunum segments from three Rab8a+/+ endolysosomal compartments, an observation made for and two Rab8a−/− mice, Fig. 6A-D). A large portion of the Paneth microvillus enzymes in Rab8a−/− enterocytes (Sato et al., 2007). cell cytoplasm is occupied by ER, where the majority of Gpr177+ We were able to examine lysosomes and multivesicular bodies gold particles were detected (Fig. 6A,B). Comparison of the (MVBs) owing to their distinctive morphologies (Fig. 6C,D; intracellular distribution of gold particles in Rab8a+/+ and Rab8a−/− supplementary material Fig. S10). We detected frequent Paneth cells indicated statistically insignificant differences in localization of Gpr177+ gold particles to lysosomes and MVBs in Gpr177 distribution in ER or non-ER compartments (Fig. 6A), Rab8a−/− but not Rab8a+/+ cells (Fig. 6D; supplementary material suggesting that Gpr177 can be exported from the ER without Rab8a. Fig. S10), suggesting that the absence of Rab8a might have caused However, we identified a significant reduction of Gpr177+ gold endolysosomal targeting of Gpr177. particles adjacent to apical or basolateral plasma membranes Western blots for endogenous Gpr177 using intestinal lysates (Fig. 6E; more Rab8a+/+ and Rab8a−/− cells are shown in showed multiple protein species, reflecting previously reported supplementary material Figs S7 and S8, respectively). This post-translational modification and alternative splicing events reduced peripheral Gpr177 localization was reflected in fewer (Tanaka et al., 2002; Yu et al., 2010, 2014a). Rab8a−/− intestines surface gold particles per Paneth cell, as well as in an ∼88% showed a reduction of the 62 kDa and 53 kDa protein species (full- reduction of particles per unit length of Paneth cell plasma length Gpr177 is predicted to be 62 kDa) (lanes 1-6, supplementary −/− ΔIEC membrane (Fig. 6E). A similar trend was observed in Rab8a material Fig. S11A). Rab8a intestines, with intact Rab8a in DEVELOPMENT

2153 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

− − Fig. 5. Rab8a deletion impairs Paneth cell maturation. (A) Lysozyme staining showed reduced numbers of lysozyme+ Paneth cells in Rab8a / crypts. − − ***P<0.001. (B) Quantitative RT-PCR showed no significant change in the Paneth cell gene expression signature in Rab8a / intestinal tissues. (C) Lysozyme − − and EdU (1 h) staining of Rab8a / organoids detected lysozyme+ Paneth cells in proliferating buds of surviving organoids. (D) TEM showed that the residual − − − − Paneth cells in Rab8a / crypts exhibit a reduction in the typical number and size of granules. Arrow points to a granule in a Rab8a / crypt. (E) An expansion of − − − − smooth ER was observed in all remaining Rab8a / Paneth cells. The thickness of stacked ER cisternae was significantly expanded in Rab8a / (834±46 nm) as compared with Rab8a+/+ (333±7 nm) Paneth cells. Scale bars: 10 µm in A; 15 µm in C; 2 µm in D; 1 µm in E. non-epithelial cells, retained the 53 kDa Gpr177 isoform (lanes 7-8, increase in targeting of GPR177-EGFP into LAMP1+ (a lysosomal supplementary material Fig. S11A), which was present in wild-type membrane protein) compartments (supplementary material MEFs but not Gpr177−/− MEFs (lanes 9-10, supplementary Fig. S11B). Treatment of RAB8A-depleted Caco2 cells with material Fig. S11A). Although these data clearly demonstrated a bafilomycin A, which blocks endolysosomal functions, increased certain impact of Rab8a loss on Gpr177 protein patterns, the mixed total GPR177 levels (supplementary material Fig. S12). Conversely, cell population in a tissue context appeared to complicate the when protein synthesis was blocked by treating RAB8A-depleted interpretation. Caco2 cells with cycloheximide, a faster reduction in the GPR177 Thus, we directly determined whether Rab8a loss increases level was observed when compared with cycloheximide-treated endolysosomal transport of Gpr177 by establishing a Caco2 cell control cells (supplementary material Fig. S13). These data line with stable RAB8A knockdown. Compared with Caco2 cells suggested that, in the absence of RAB8A-mediated exocytosis, treated with non-specific shRNA, RAB8A depletion caused a 3-fold there is an enhanced endolysosomal transport and clearance DEVELOPMENT

2154 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

Fig. 6. Rab8a deletion affects Gpr177 transport to the plasma membrane but not ER export. (A) Quantification of Gpr177+ immunogold particle distributions − − − − in the ER, non-ER Golgi/vesicle and plasma membrane of Rab8a+/+ and Rab8a / Paneth cells (particles counted from 12 Rab8a+/+ and nine Rab8a / Paneth − − cells from three Rab8a+/+ and two Rab8a / mice, respectively). (B) The majority of Gpr177+ immunogold particles (arrows) were detected in the ER of Rab8a+/+ − − and Rab8a / Paneth cells. (C,D) Representative micrographs showing Gpr177+ immunogold particles in Golgi and lysosomes. Gold particles were detected in − − lysosomes (open arrowheads) in Rab8a / (D) but not in Rab8a+/+ (C). (E) Gpr177+ immunogold particles were frequently detected at apical or basolateral plasma − − membranes in wild-type Paneth cells. A significant reduction in plasma membrane-localized Gpr177+ particles was detected in Rab8a / Paneth cells per unit length of plasma membrane. Data were collected for a total length of 312 µm plasma membrane in 12 Rab8a+/+ Paneth cells and 328 µm plasma membrane in 14 − − − − Rab8a / Paneth cells from three Rab8a+/+ and two Rab8a / mice, respectively. Arrowheads point to plasma membranes between a stem cell and a Paneth cell. lu, lumen. *P<0.05. Scale bars: 500 nm. of GPR177, yet RAB8A-depleted cells continue to make new derived Rab8a−/−;Lgr5EGFP−IRES−CreERT2/+ mice, and analyzed GPR177. proliferative Lgr5+ stem cells by EGFP and EdU labeling. Increased EdU+ Lgr5+ cells were detected in both the small intestine and colon − − − − Rab8a deletion induces changes of crypt cell organization of Rab8a / ;Lgr5EGFP IRES CreERT2/+ mice (Fig. 7A). Total Rab8a-deficient mice provided an opportunity to examine the crypt numbers of Lgr5+ cells per crypt also increased in Rab8a−/− response to weakened Wnt production at the stem cell niche. We intestines (Fig. 7A). The numbers of transit amplifying cells first examined the fast-cycling crypt base columnar (CBC) cells in and mitotic crypt cells (phosphorylated histone H3+) were −/− −/− Rab8a mice using Lgr5 as an indicator (Barker et al., 2007). We also increased in Rab8a crypts (supplementary material DEVELOPMENT

2155 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

−/− Fig. 7. Adaptive changes in Rab8a crypts. (A) EdU labeling (1 and 3 h, red) of mouse small intestine showed a significant increase in proliferative Lgr5+ cells − − − − − − in Rab8a / ;Lgr5EGFP IRES CreERT2 mice. The total number of Lgr5+ cells was also increased in Rab8a / crypts. Around 45 crypts that contained Lgr5+ cells were − − − − − − analyzed in each tissue section of independent Lgr5EGFP IRES CreERT2 or Rab8a / ;Lgr5EGFP IRES CreERT2 mice. (B) EdU labeling of the colon showed an − − − − increase in proliferative Lgr5+ cells in Rab8a / ;Lgr5EGFP IRES CreERT2 mice. (C) Co-staining of BrdU (1 h, brown) and β-gal (blue) showed increased BrdU+ cells − − in Axin2lacZ/+;Rab8a / intestines, despite decreased Axin2 reporter activity in the same crypts. One hundred continuous crypts were analyzed in each section of − − − − independent Rab8a+/+ and Rab8a / mice. (D) Rab8a / intestines contained more cells with the strongest level of Hopx immunoreactivity (arrows). Arrowheads point to cells with moderately higher immunoreactivity than in wild type. One hundred continuous crypts were analyzed in each section of independent Rab8a+/+ − − − − and Rab8a / mice. (E) Quantitative RT-PCR detected increased Wnt3a levels in Rab8a / intestines (n=3 for each genotype). (F,G) RNA in situ hybridization to − − detect Wnt3 (F) or Wnt3a (G), showing ectopic activation of Wnt3a in Rab8a / crypts, whereas Wnt3 was largely unaffected. *P<0.05, ***P<0.001. Scale bars: 10 µm.

Fig. S14; data not shown), which might explain the increased c-Myc cells did not appear to be impacted by weakened Wnt signaling – an in Rab8a−/− intestines (Fig. 3B). Co-staining for β-gal and BrdU unexpected observation that contrasts with the drastic loss of using Rab8a−/−;Axin2lacZ/+ tissues identified a 2-fold increase in Rab8a−/− organoids in vitro (see Discussion). BrdU+ cells in Rab8a−/− crypts, where reduced β-gal activity was Crypt cells demonstrate great plasticity (Tetteh et al., 2014). The −/− seen (Fig. 7C). Thus, the in vivo proliferation of Rab8a crypt tissue regenerative program could be activated by injury in a numberof DEVELOPMENT

2156 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046 slow-cycling or secretory precursor cell types (Buczacki et al., 2013; organoids. Third, an independent screening assay by Demir et al. Powell et al., 2012; Tian et al., 2011; van Es et al., 2012; Yan et al., showed that RAB8A does not affect Wnt reception (Demir et al., 2012). Bmi1+ cells have been shown to be relatively resistant to Wnt 2013). These data corroborated the view that, unlike RAB8B which signaling perturbation (Yan et al., 2012). The reduced Wnt regulates Wnt reception and LRP6 endocytosis, RAB8A plays a concentration in Rab8a−/− crypts might stress crypt cells into certain distinct role in Wnt-producing cells. Accordingly, we did not observe adaptive changes. Using quantitative RT-PCR we surveyed several an impact of RAB8B on Wnt secretion. ‘quiescent’ stem cell markers but only detected an increase in the Bmi1 In Rab8a−/− Paneth cells or subepithelial fibroblasts, immunogold mRNA level (supplementary material Fig. S15). Using a Hopx labeling of Gpr177 and surface protein isolation demonstrated reduced antibody, we also analyzed the reported +4 cell type that has been localization of Gpr177 adjacent to the cell surface. It appears that shown by lineage tracing to convert to Lgr5+ cells (Takeda et al., disruption of Rab8a-dependent transport weakens Gpr177-mediated 2011). Rab8a+/+ intestinal epithelia contained about three Hopx+ cells Wnt secretion by rerouting Gpr177 into endolysosomal compartments (experimentally defined as strongly immunopositive for nuclear (Fig. 8). A similar observation has been made following defective Hopx) in every ten adjacent crypts, whereas Rab8a−/− intestines retromer-dependent Gpr177 retrieval (Eaton, 2008). After synthesis in showed about seven or eight Hopx+ cells in the same number of crypts the ER, lysosomal hydrolases and membrane proteins are transported (Fig. 7D). Rab8a−/− intestines also contained multiple Hopx+ cells directly from the Golgi apparatus and trans-Golgi network to late in a single crypt, with generally stronger nuclear immunoreactivity endosomes via vesicles that exclude other proteins intended for compared with Rab8a+/+ intestines (Fig. 7D, arrows and arrowheads distinct destinations. Possibly, loss of Rab8a causes mispackaging indicate strong and moderate staining, respectively). Hopx+ cells of Gpr177 into lysosome-targeting vesicles. However, it is still were found in Rab8a−/− epithelium at positions other than +4 unclear how loss of Rab8a would trigger Gpr177 vesicular flow into (Fig. 7D). One-hour EdU labeling of Rab8a−/− miceledtopositive lysosome-targeting vesicles. In the absence of Rab8a, post-Golgi labeling of some Hopx+ cells, whereas this was rarely found Gpr177 vesicles might be hijacked by a distinct Rab GTPase (e.g. in Rab8a+/+ crypts (supplementary material Fig. S16A). Rab9), facilitating endolysosomal fusion. Brush border enzymes are Co-immunofluorescence analyses of Lgr5+ cells and Hopx+ cells also mistargeted into lysosomes in Rab8a−/− enterocytes (Sato et al., in Rab8a−/−;Lgr5EGFP−IRES−CreERT2/+ intestines showed increased 2007), hinting at some similarities in defective cargo transport in the numbers of both cell types (supplementary material Fig. S16B). absence of Rab8a. In different cell types we observed differential Wnt3, but not Wnt3a, is normally expressed in Paneth cells of influences of Rab8a loss on the Gpr177 protein pattern. This was the adult mouse intestine (Farin et al., 2012; Gregorieff et al., 2005). probably attributable to cell-specific trafficking factors. Thus, the When we conducted Wnt ligand expression analyses, we molecular basis of Rab8a-mediated Gpr177 traffic remains to be unexpectedly detected ectopic Wnt3a mRNA in Rab8a−/− explored in detail. Of note, Sec4, the Rab8a homolog in yeast, intestinal crypts (Fig. 7E). By in situ hybridization with Wnt3- and predominantly mediates post-Golgi vesicle secretion (Das and Guo, Wnt3a-specific probes (Roelink and Nusse, 1991), ectopic activation 2011; Hutagalung and Novick, 2011; Seabra et al., 2002). of Wnt3a was confirmed in Rab8a−/− crypts (Fig. 7G), whereas Wnt3 The view that Rab8a vesicles promote the export of Gpr177-Wnt was largely unaffected (Fig. 7F). This ectopic Wnt3a expression also provided additional support for the physiological involvement might suggest a cell type change rather than increased Wnt3a of Rab8 in GPCR anterograde trafficking (Wang and Wu, 2012). secretion from these cells, as Rab8a−/− cells failed to properly secrete RAB8 was shown to modulate the exocytosis of metabotrophic Wnt3a-Gluc (Fig. 2C). Together, these data suggested that, in the glutamate receptor (Esseltine et al., 2012) and α2B and β2 absence of Rab8a, the weakened Wnt signaling blocked Paneth cell adrenergic receptors (Dong et al., 2010) in human embryonic maturation from Lgr5+ secretory precursors and that this might have initiated an aberrant differentiation program.

DISCUSSION Our study of Rab8a function in the mouse intestinal crypt compartment extended previous analyses performed in differentiated enterocytes (Sato et al., 2007). The crypt defects observed in Rab8a knockout mice are likely to precede the apical transport abnormalities in enterocytes that are derived from the crypt progenitors. Our finding that Rab8a affects Wnt secretion and Paneth cell maturation impinges on a growing body of evidence linking vesicular traffic to niche signal transduction and maintenance (Feng and Gao, 2014; Goldenring, 2013).

Wnt secretion Rab8a deletion clearly impacted Wnt signaling in the intestinal crypts. This was strongly supported by Axin2 reporter analysis in vivo and in organoid culture, as well as by the defective Paneth cell maturation in Rab8a−/− crypts. Mechanistically, these phenotypes could be caused by defective traffic of Wnt receptors such as Fzd or Lrp6. This possibility was ruled out on the basis of several lines of evidence. First, Rab8a−/− MEFs still responded to exogenous Wnt Fig. 8. Rab8a facilitates anterograde transport of Gpr177-Wnt. Rab8a facilitates the transport of post-Golgi Gpr177-Wnt vesicles to the plasma ligands by phosphorylating the Lrp6 cytosolic tail and activating a membrane for secretion. Loss of Rab8a attenuates Gpr177 exocytotic traffic Topflash Wnt reporter. Second, exogenous Wnt3a and Gsk3β and, via an unknown mechanism (red dotted arrows), may reroute Gpr177 into −/− inhibitor partially restored the growth and survival of Rab8a endolysosomes. TGN, trans-Golgi network. DEVELOPMENT

2157 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046 kidney cells and primary neurons, with an interaction with the Lgr5+ cells are certainly different from typical CBCs in healthy C-terminus of β2 adrenergic receptors (Dong et al., 2010). animals, and might reflect Paneth cell precursors, as they expressed In Rab8a−/− MEFs, we observed a rather selective impact on Wnt Paneth cell genes. Third, the proposed quiescent crypt cells, in secretion. Rabs and other membrane trafficking regulators are often particular those that are Bmi1+, have been shown to be resistant to considered to be generic modulators of protein transport. However, Wnt perturbation and capable of converting into Lgr5+ cells (Tian accumulating evidence suggests that perturbing these trafficking et al., 2011; Yan et al., 2012). Rab8a−/− crypts also contained more processes impacts a rather specific pathway (Knowles et al., 2015; Hopx+ cells, in addition to increased Lgr5+ cells, collectively Yu et al., 2014c). For instance, depletion of RAB8B impacted Wnt suggesting a certain degree of crypt cell repopulation. signalosome activity but not other signaling pathways (Demir et al., Although intestinal crypt-villus damage was induced in extreme 2013). Disrupting retromer function by ablating Vps35 or Snx3 Wnt perturbation models, as exemplified by dickkopf homolog 1 impaired primarily Wnt secretion (Eaton, 2008; Harterink et al., (Dkk1) overexpression (Kuhnert et al., 2004; Pinto et al., 2003) and 2011). Furthermore, partial loss of Rab1 and Rab11 preferentially Tcf4−/− mice (van de Wetering et al., 2002), removal of Wnt affected Notch signaling in Drosophila (Charng et al., 2014; Emery production from a number of Wnt-producing sources has so far been et al., 2005), whereas genetic mutation of Rab23 specifically insufficient to perturb crypt stem cells (Farin et al., 2012; Kabiri et al., perturbed the Hedgehog signaling pathway (Eggenschwiler et al., 2014; San Roman et al., 2014). Factors contributing to intestinal 2001; Evans et al., 2003), with its potential targets being Suppressor regenerative abilities may arise from multiple sources. Injury-induced of fused (Chi et al., 2012) or/and Smoothened – another GPCR apoptotic cells have been shown to be a source of Wnt3 to support (Boehlke et al., 2010). As individual vesicles contain heterogeneous regeneration (Galliot, 2013). Autocrine Wnts produced by adult stem cargoes, it is unlikely that Rab8a only traffics Gpr177. cells may constitute their self-niche for renewal (Lim et al., 2013). Finally, crypt cells that are insensitive to either R-spondin or Dkk1 Paneth cells may exist to replenish the epithelia (Yan et al., 2012), presumably in a The organoid-forming capacity of Lgr5+ stem cells in culture is Wnt-independent fashion. Given that the vast majority of colon enhanced byassociation with Paneth cells (Sato et al., 2011). We found cancer cells contain a constitutively active Wnt pathway and thereby that Rab8a−/− organoids resemble Wnt3−/− or Atoh1−/− organoids do not rely on external Wnts, it is not entirely surprising to observe (Durand et al., 2012; Farin et al., 2012), showing poor clonogenic strong cell resilience to Wnt ligand perturbation. Cell lineage analysis activities. The vast majority of surviving Rab8a organoids contained might help us to better understand the cellular adaptation in Rab8a- tiny Paneth cell-containing buds, suggesting that the few remaining deficient crypt. Whether Gpr177 regulates Wnt secretion in intestinal Paneth cells facilitated residual clonogenic activities of the knockout stem cells in addition to Paneth cells, as it does in other organs (Jiang organoids. These data favored the notion that Paneth cells constituted et al., 2013; Stefater et al., 2011), requires further study. Taken the major stem cell supporters in cultured enteroids after growth together, Rab8a deletion induces epithelial stress that provokes crypt support from non-epithelial compartments was eliminated. When cell alteration. We conclude that Rab8a affects Wnt secretion and Rab8a−/− organoids were treated with Porcupine inhibitor to deplete Paneth cell maturation at the intestinal stem cell niche. residual Wnt production, their growth was further arrested, indicating that Rab8a deletion did not completely abrogate Wnt secretion. MATERIALS AND METHODS In vivo, most Rab8a−/− intestinal crypts were largely devoid of Mice and cells −/− ΔIEC EGFP−IRES−CreERT2 fl/fl lacZ/+ lysozyme+ Paneth cells. However, the transcriptional level of Rab8a , Rab8a , Lgr5 , Gpr177 and Axin2 micehave been described previously (Barkeret al., 2007; Fu et al., 2011; Lustig multiple Paneth cell-specific genes remained normal. This − − Δ et al., 2002; Sato et al., 2007). Rab8a / and Rab8a IEC mice die at ∼4 weeks; suggested that the maturation of Paneth cells, rather than Paneth −/− all comparisons were made between littermates of at least three mice for each cell fate commitment, was affected in Rab8a crypts. Our model genotype unless stated otherwise. All experiments were repeated two or three represents an independent genetic setting for the evaluation of the times; only consistent results are presented. Experimental procedures were crypt response to Paneth cell defects. approved by Rutgers University Institutional Animal Care and Use Committee. A subset of Lgr5+ LRCs was identified to express both CBC and Procedures for derivation of Rab8a−/−, Rab8b knockdown, Gpr177−/− MEFs Paneth cell gene signatures (Buczacki et al., 2013) and proposed to and intestinal organoids are detailed in the supplementary material Methods. constitute precursorsof Paneth cells capable of regeneration following Phenotypic analyses of cells and tissues have been described previously (Gao injury. Our data suggested that the lack of extracellular Wnts in et al., 2009; Sakamori et al., 2012; Yu et al., 2014b). Rab8a−/− crypts might resemble injury-like stress that blocked Paneth cell maturation from their precursors. Importantly, despite the lack of Gpr177 vesicle tracking MEFs were transfected with Gpr177-mCherry using Lipofectamine 2000 mature Paneth cells, Wnt3-expressing crypt cells remained in Rab8a (Life Technologies) as per manufacturer’s specifications. Forty-eight hours knockout crypts, in addition to some crypt cells ectopicallyexpressing after transfection, cells were trypsinized and seeded into a 35 mm glass-bottom Wnt3a, suggesting that aberrant transdifferentiation might be dish (MatTek Corporation) and imaged after 24 h. Images were acquired using triggered by stress induced by Rab8a deletion. a Zeiss Axio Observer Z1 equipped with a 100× objective, heated stage and CO2 controller. Frames were captured at the maximum possible speed Stem cells allowing z-plane correction using a definite focus system (Zeiss). Data were This study provided an opportunity to examine the intestinal analyzed using ImageJ (NIH). Images were inverted and vesiclestracked using the MTrackJ plug-in (Meijering et al., 2012). Sixteen Gpr177-mCherry stem cell response to a perturbed niche Wnt signal. First, Rab8a −/− deficiency reduced Wnt secretion by ∼80%. Cells might respond to vesicles were tracked in control and Rab8a MEFs and data were collected this by increasing their sensitivity to the ligand. This was indeed from two independent experiments. Speeds were plotted as frequency distribution (percentage of total steps) with a bin size of 70 nm/s. observed in Rab8a−/− MEFs. Second, LRC secretory precursors express both Lgr5 and Paneth cell gene signatures and contribute to Dual-luciferase assay and fluorescent SEAP assay epithelial regeneration upon tissue injury (Buczacki et al., 2013). To measure Wnt3a-Gluc, ShhNRen and Metridia luciferase secretions, We found a clear blockage of Paneth cell maturation and an increase MEFs of various genotypes were transiently co-transfected with Wnt3a- + −/− −/− in Lgr5 cell number in Rab8a crypts. These increased Rab8a Gluc, ShhN-Ren or Met-Luc, with firefly luciferase serving as transfection DEVELOPMENT

2158 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

control. Twenty-four hours after transfection, supernatants and cell lysates K4Fe(CN)6, 2 mM MgCl2, 0.01% sodium deoxycholate, 0.02% NP40, were collected and subjected to dual-luciferase assay (Promega, E1980) 1 mg/ml X-Gal (Fisher Scientific, 50-213-181) overnight at 37°C. The using the Glomax multidetection system (Promega). Each reaction consisted samples were rinsed with PBS and mounted for imaging with a Nikon of 50 μl medium or 10 μl cell lysate, and 50 μl luciferase assay substrate and TE2000 microscope. Data represent three independent experiments. 50 μl of Stop&Glo Reagent. Procedures for immunofluorescence, immunohistochemistry and in situ The SEAP assay was performed using the Great EscAPe Fluorescence hybridization have been described previously (Gao and Kaestner, 2010; Gao Detection Kit (Clontech, 631704). 25 μl supernatant or cell lysate was et al., 2009; Sakamori et al., 2012). A complete list of antibodies is provided diluted with the same volume of 1×dilution buffer in a 96-well plate, mixed in the supplementary material Methods. Wnt3a and Wnt3 RNA in situ gently on a rotating platform for 5 min at room temperature and incubated at probes were as described (Gregorieff et al., 2005; Roelink and Nusse, 1991). 65°C for 30 min. Samples were then cooled to room temperature; 97 μl assay Quantitative RT-PCR is described in the supplementary material Methods buffer was added and incubated for 5 min at room temperature; 3 μl1mM and Table S1. MUP (substrate) was added and incubated with the sample at room temperature for 1 h in the dark. Fluorescent units were read using the Biotinylation assay Glomax multidetection system. Luminescence units from Gaussia Cells were grown to ∼90% confluence in 10-cm dishes, washed three times luciferase, Renilla luciferase and Metridia luciferase were normalized with cold 1× PBS and incubated with 10 ml cold Biotin solution with gentle against intracellular firefly luciferase. Assays were repeated three or more rocking at 4°C. Biotinylation was quenched after 30 min and cell lysates times with three to six technical replicates per cell line each time. collected as per manufacturer’s instructions (Pierce, 89881). Then, 300 μg cell lysates was added to 100 μl Neutravidin beads (Pierce) for 1 h at room Topflash reporter assay temperature, washed four times with the wash buffer provided and eluted in To compare Wnt3a responsiveness in wild-type, Rab8a−/− and Rab8b 4× LDS containing 50 mM DTT at 70°C for 15 min. The supernatant was knockdown MEFs, cells were co-transfected with Topflash and Renilla collected and immunoblot analysis performed using anti-Frizzled (1-10), luciferase for 24 h. The cells were then serum starved for 3 h in Wnt-free anti-Lrp6, anti-phospho-Lrp6 (Ser1490), anti-N-cadherin and anti-Gpr177 DMEM and treated with 20ng/ml recombinant murine Wnt3a (Peprotech, 315- (see supplementary material Methods). Data represent three independent 20). After 5 h medium was removed, cells washed with 1× PBS, lysed and experiments. luciferase activity detected using the dual-luciferase assay and Glomax system (Promega). To compare Wnt5a secretory abilities by wild-type and Rab8a−/− Wnt5a/b secretion assay MEFs, cells were simultaneously transfected with pcDNA-WNT5A, Topflash MEFs were grown to ∼90% confluence in 10-cm dishes, washed three times andRenilla luciferase inWnt-free lactalbumin hydrolysate(SAFC Biosciences, with 1× PBS and cultured in 10 ml 1× lactalbumin hydrolysate (SAFC 58901-C) in DMEM for 16-18 h. Topflash activity was detected in cell lysates Biosciences, 58901-C) in DMEM. After 24 h, medium was collected in and normalized to intracellular Renilla luciferase. Data represent three 50 ml Falcon tubes and centrifuged at 10,000 g for 10 min. Supernatant was independent experiments with comparable transfection efficiencies. loaded onto an Amicon Ultra-15 centrifugal filter system (Millipore, UFC 901024, 10K MWCO) and centrifuged at 5000 g at room temperature for Co-immunoprecipitation, western blot and GST pull-down 1 h. The concentrate (∼100 μl) was collected and subjected to immunoblot Immunoprecipitations were performed with anti-Flag M2 affinity gel analysis using anti-Wnt5a/b and anti-histone H3 (see supplementary (Sigma, A2220) using 2 mg total lysates extracted from 3×Flag-GPR177 material Methods). stable HeLa cells. Reactions were incubated for 8 h at 4°C, washed three times with buffer containing 1% Triton X-100 and eluted with 40 μl 3×Flag Transmission electron microscopy (TEM) analysis and Gpr177 peptides. Western blot and GST pull-down procedures have been described immunogold labeling previously (Gao and Kaestner, 2010; Gao et al., 2009; Sakamori et al., TEM procedures have been described previously (Gao et al., 2011; 2012). The wash buffer comprised 50 mM Tris HCl (pH 7.5), 150 mM Sakamori et al., 2012). For Gpr177 immunogold labeling and TEM NaCl, 1 mM EDTA and 1% Triton X-100. analysis, duodenal and jejunal tissues were dissected from wild-type (n=3) For GST pull-down, GST, GST-RAB8A, GST-CDC42 and GST-JFC-D1 and Rab8a−/− (n=2) mice and immediately fixed as ∼1 mm fragments in fusion proteins were expressed in BL21 cells induced by 0.5 mM IPTG and 2.5% paraformaldehyde in cacodylate buffer (pH 7.4) overnight. The tissue cultured at room temperature for 18 h (Sakamori et al., 2012). The bacterial was sliced to 100-200 μm thickness on a Vibratome and frozen between two cells were then resuspended in 2 ml 1× PBS with 1% Triton X-100, 0.1 mg/ brass ‘top hats’ in a HPM010 (Abra Fluid) at 5000 psi at –180°C. Next, the ml lysozyme (Sigma, L6876), 1 mM PMSF and 1× bacterial protease frozen tissue was transferred to frozen glass-distilled 100% acetone and inhibitors (Sigma, P8465), incubated on ice for 30 min followed by dehydrated at –90°C for 48 h. The tissue was then infiltrated with HM-20 sonication. GST protein-containing lysates thus collected were incubated lowicryl (Electron Microscopy Sciences) and polymerized with 360 nm with pre-swollen glutathione-agarose beads (Molecular Probes, G-2878) at light at –50°C in a dry nitrogen environment. Tissue sections (60 nm thick) 4°C for 1 h. The beads were washed three times with 1× PBS. To check containing Paneth cells were immunolabeled with Gpr177 antibody (Fu GST-protein expression, a portion of GST-protein-conjugated beads was et al., 2009) at 1:250 to 1:50 in 5% BSA, 0.1% cold water fish gelatin in PBS denatured in 4× LDS (Life Technologies, NP0007) at 70°C for 15 min and (pH 7.4). No-primary-antibody control and Gpr177-deficient cells were subjected to SDS-PAGE followed by Coomassie Blue staining (Invitrogen, used in initial tests to optimize labeling conditions. Images shown were LC6060). Comparable amounts of beads were incubated with 1 mg cell immunolabeled with 1:100 primary antibody. Stable antigen-antibody lysates from 3×Flag-GPR177 stable HeLa cells for 1 h at 4°C. Beads were complexes were detected with protein A conjugated to either 15 nm gold washed with PBS containing 1% Triton X-100 and mammalian protease colloids or 20 nm gold colloids (Electron Microscopy Sciences). Imaging inhibitors (Roche, 11 697 498 001), denatured in 4× LDS at 70°C for 15 was performed with an FEI Tecnai-12 microscope at 80 keV using a nominal min, and subjected to anti-Flag western blot analysis. Data represent three magnification of 6500×. Montage images were collected using serial EM independent experiments. and stitched together with the IMOD subroutine Blendmont (Kremer et al., 1996; Mastronarde, 2005). Each image has a pixel dimension of ∼3 nm, β-galactosidase staining, confocal immunofluorescence, such that each spherical gold particle should fill five pixels and resizing of immunohistochemistry, RNA in situ hybridization and the images should provide information that reveals the approximate volume quantitative RT-PCR the gold would occupy in the images. Immunogold particles were counted For β-galactosidase staining, tissue sections or whole-mount organoids in manually, excluding particles within the nuclear area. Area and perimeter of chamber slides were rinsed with 1× PBS, fixed in fixative (1% individual Paneth cells were measured using Photoshop CS (Adobe). formaldehyde, 0.2% glutaraldehyde, 2 mM MgCl2, 5 mM EGTA and Numbers of immunogold particles in each subcellular compartment per unit 0.02% NP40) for 15 min, washed in PBS three times at room temperature, area (for ER and Golgi) or per unit length (for plasma membrane) were and stained with staining solution comprising 5 mM K3Fe(CN)6,5mM calculated and compared by t-test. Data were collected from three DEVELOPMENT

2159 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046 independent labeling experiments, performed twice on two pairs of Buczacki, S. J. A., Zecchini, H. I., Nicholson, A. M., Russell, R., Vermeulen, L., independent wild-type and knockout tissues, and once on a wild-type Kemp, R. and Winton, D. J. (2013). Intestinal label-retaining cells are secretory mouse tissue. precursors expressing Lgr5. Nature 495, 65-69. Cadigan, K. M. and Peifer, M. (2009). Wnt signaling from development to disease: insights from model systems. Cold Spring Harb. Perspect. Biol. 1, a002881. Quantifications and statistical analyses Charng, W.-L., Yamamoto, S., Jaiswal, M., Bayat, V., Xiong, B., Zhang, K., Quantifications were performed using either two-tailed, unpaired or paired Sandoval, H., David, G., Gibbs, S., Lu, H.-C. et al. (2014). Drosophila Tempura, one-tailed Student’s t-test on the basis of experimental setups. Mean values a novel protein prenyltransferase alpha subunit, regulates notch signaling via are shown, with error bars representing s.e.m. Western blots were quantified Rab1 and Rab11. PLoS Biol. 12, e1001777. β Chen, B., Dodge, M. E., Tang, W., Lu, J., Ma, Z., Fan, C.-W., Wei, S., Hao, W., using ImageJ. -gal staining areas and intensities were measured and Kilgore, J., Williams, N. S. et al. (2009). Small molecule-mediated disruption of quantified with the ‘color deconvolution’ plug-in provided in ImageJ. Non- Wnt-dependent signaling in tissue regeneration and cancer. Nat. Chem. Biol. 5, linear regression analysis was used for curve fitting of the frequency of 100-107. instantaneous speeds of GPR177-mCherry vesicles in MEFs. The Chi, S., Xie, G., Liu, H., Chen, K., Zhang, X., Li, C. and Xie, J. (2012). Rab23 histograms were fitted to a single Lorentzian distribution. All graphs were negatively regulates Gli1 transcriptional factor in a Su(Fu)-dependent manner. Cell Signal. 24, 1222-1228. constructed using GraphPad Prism 5. Ching, W. and Nusse, R. (2006). A dedicated Wnt secretion factor. Cell 125, 432-433. Acknowledgements Clarke, A. R. (2006). Wnt signalling in the mouse intestine. Oncogene 25, We thank Dr Wei Guo (University of Pennsylvania) for the GST-RAB8A plasmid and 7512-7521. Dr Lawrence Lum (University of Texas Southwestern Medical Center) for the Wnt3a- Clevers, H. (2006). Wnt/beta-catenin signaling in development and disease. Cell Gluc plasmid. 127, 469-480. Clevers, H. and Nusse, R. (2012). Wnt/β-catenin signaling and disease. Cell 149, Competing interests 1192-1205. The authors declare no competing or financial interests. Clevers, H., Loh, K. M. and Nusse, R. (2014). An integral program for tissue renewal and regeneration: Wnt signaling and stem cell control. Science 346, 1248012. Author contributions Das, A. and Guo, W. (2011). Rabs and the exocyst in ciliogenesis, tubulogenesis S.D., A.R., R.D., A.Ha., W.H., E.M.B., M.P.V. and N.G. conceived the project; S.D., and beyond. Trends Cell Biol. 21, 383-386. S.Y., R.S., P.V., Q.F., J.Fl., A.H., J.F., E.S., E.M.B. and N.G. performed experiments Das, S., Yu, S., Sakamori, R., Stypulkowski, E. and Gao, N. (2012). Wntless in and analyzed the data; S.D., E.M.B., M.P.V. and N.G. wrote the paper. Wnt secretion: molecular, cellular and genetic aspects. Front. Biol. (Beijing) 7, 587-593. Funding de Groot, R. E. A., Farin, H. F., Macůrková, M., van Es, J. H., Clevers, H. C. and This work was supported by National Institutes of Health (NIH) grants [DK102934, Korswagen, H. C. (2013). Retromer dependent recycling of the Wnt secretion DK085194, DK093809, CA178599], Charles and Johanna Busch Memorial Award factor Wls is dispensable for stem cell maintenance in the mammalian intestinal [659160] and National Science Foundation (NSF)/BIO/IDBR [1353890] to N.G.; NIH epithelium. PLoS ONE 8, e76971. grants [DK088868, DK099251] to M.P.V.; RU Faculty Research Grant [281708] to de Lau, W., Barker, N. and Clevers, H. (2007). WNT signaling in the normal M.P.V. and N.G.; NIH grants [CA106308, DE015654] to W.H.; American Federation intestine and colorectal cancer. Front. Biosci. 12, 471-491. Demir, K., Kirsch, N., Beretta, C. A., Erdmann, G., Ingelfinger, D., Moro, E., for Aging Research (AFAR) [#RAG13447] to R.D. N.G. is supported by a Research Argenton, F., Carl, M., Niehrs, C. and Boutros, M. (2013). RAB8B is required for Scholar Grant [RSG-15-060-01-TBE] from the American Cancer Society. S.Y. activity and caveolar endocytosis of LRP6. Cell Rep. 4, 1224-1234. received a New Jersey Commission on Cancer Research Postdoctoral Fellowship Dong, C., Yang, L., Zhang, X., Gu, H., Lam, M. L., Claycomb, W. C., Xia, H. and [DFHS13PPC016]. Deposited in PMC for immediate release. Wu, G. (2010). Rab8 interacts with distinct motifs in alpha2B- and beta2- adrenergic receptors and differentially modulates their transport. J. Biol. Chem. Supplementary material 285, 20369-20380. Supplementary material available online at Durand, A., Donahue, B., Peignon, G., Letourneur, F., Cagnard, N., Slomianny, C., http://dev.biologists.org/lookup/suppl/doi:10.1242/dev.121046/-/DC1 Perret, C., Shroyer, N. F. and Romagnolo, B. (2012). Functional intestinal stem cells after Paneth cell ablation induced by the loss of transcription factor Math1 References (Atoh1). Proc. Natl. Acad. Sci. USA 109, 8965-8970. Ang, A. L., Fölsch, H., Koivisto, U.-M., Pypaert, M. and Mellman, I. (2003). The Eaton, S. (2008). Retromer retrieves wntless. Dev. Cell 14, 4-6. Rab8 GTPase selectively regulates AP-1B-dependent basolateral transport in Eaton, S., Wepf, R. and Simons, K. (1996). Roles for Rac1 and Cdc42 in planar polarized Madin-Darby canine kidney cells. J. Cell Biol. 163, 339-350. polarization and hair outgrowth in the wing of Drosophila. J. Cell Biol. 135, Angers, S. and Moon, R. T. (2009). Proximal events in Wnt signal transduction. Nat. 1277-1289. Eggenschwiler, J. T., Espinoza, E. and Anderson, K. V. (2001). Rab23 is an Rev. Mol. Cell Biol. 10, 468-477. essential negative regulator of the mouse Sonic hedgehog signalling pathway. Bänziger, C., Soldini, D., Schütt, C., Zipperlen, P., Hausmann, G. and Basler, K. Nature 412, 194-198. (2006). Wntless, a conserved membrane protein dedicated to the secretion of Wnt Emery, G., Hutterer, A., Berdnik, D., Mayer, B., Wirtz-Peitz, F., Gaitan, M. G. and proteins from signaling cells. Cell 125, 509-522. Knoblich, J. A. (2005). Asymmetric Rab11 endosomes regulate delta recycling Barker, N., van Es, J. H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen, M., and specify cell fate in the Drosophila nervous system. Cell 122, 763-773. Haegebarth, A., Korving, J., Begthel, H., Peters, P. J. et al. (2007). Esseltine, J. L., Ribeiro, F. M. and Ferguson, S. S. G. (2012). Rab8 modulates Identification of stem cells in small intestine and colon by marker gene Lgr5. metabotropic glutamate receptor subtype 1 intracellular trafficking and signaling in Nature 449, 1003-1007. a protein kinase C-dependent manner. J. Neurosci. 32, 16933-16942. Bartscherer, K., Pelte, N., Ingelfinger, D. and Boutros, M. (2006). Secretion Evans, T. M., Ferguson, C., Wainwright, B. J., Parton, R. G. and Wicking, C. of Wnt ligands requires Evi, a conserved transmembrane protein. Cell 125, (2003). Rab23, a negative regulator of hedgehog signaling, localizes to the 523-533. plasma membrane and the endocytic pathway. Traffic 4, 869-884. Belenkaya, T. Y., Wu, Y., Tang, X., Zhou, B., Cheng, L., Sharma, Y. V., Yan, D., Fanto, M., Weber, U., Strutt, D. I. and Mlodzik, M. (2000). Nuclear signaling by Rac Selva, E. M. and Lin, X. (2008). The retromer complex influences Wnt secretion and Rho GTPases is required in the establishment of epithelial planar polarity in by recycling wntless from endosomes to the trans-Golgi network. Dev. Cell 14, the Drosophila eye. Curr. Biol. 10, 979-988. 120-131. Farin, H. F., Van Es, J. H. and Clevers, H. (2012). Redundant sources of Wnt Bilic, J., Huang, Y.-L., Davidson, G., Zimmermann, T., Cruciat, C.-M., Bienz, M. regulate intestinal stem cells and promote formation of Paneth cells. and Niehrs, C. (2007). Wnt induces LRP6 signalosomes and promotes Gastroenterology 143, 1518-1529.e7. dishevelled-dependent LRP6 phosphorylation. Science 316, 1619-1622. Feng, Q. and Gao, N. (2014). Keeping Wnt signalosome in check by vesicular Boehlke, C., Bashkurov, M., Buescher, A., Krick, T., John, A.-K., Nitschke, R., traffic. J. Cell. Physiol. 230, 1170-1180. Walz, G. and Kuehn, E. W. (2010). Differential role of Rab proteins in ciliary Franch-Marro, X., Wendler, F., Guidato, S., Griffith, J., Baena-Lopez, A., Itasaki, trafficking: Rab23 regulates smoothened levels. J. Cell Sci. 123, 1460-1467. N., Maurice, M. M. and Vincent, J.-P. (2008). Wingless secretion requires Boutros, M. and Mlodzik, M. (1999). Dishevelled: at the crossroads of divergent endosome-to-Golgi retrieval of Wntless/Evi/Sprinter by the retromer complex. Nat. intracellular signaling pathways. Mech. Dev. 83, 27-37. Cell Biol. 10, 170-177. Bryant, D. M., Datta, A., Rodrıguez-Fraticelli,́ A. E., Peränen, J., Martın-́ Fu, J., Jiang, M., Mirando, A. J., Yu, H.-M. I. and Hsu, W. (2009). Reciprocal Belmonte, F. and Mostov, K. E. (2010). A molecular network for de novo regulation of Wnt and Gpr177/mouse Wntless is required for embryonic axis

generation of the apical surface and lumen. Nat. Cell Biol. 12, 1035-1045. formation. Proc. Natl. Acad. Sci. USA 106, 18598-18603. DEVELOPMENT

2160 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

Fu, J., Ivy Yu, H.-M., Maruyama, T., Mirando, A. J. and Hsu, W. (2011). Gpr177/ Kim, T.-H., Escudero, S. and Shivdasani, R. A. (2012). Intact function of Lgr5 mouse Wntless is essential for Wnt-mediated craniofacial and brain development. receptor-expressing intestinal stem cells in the absence of Paneth cells. Proc. Dev. Dyn. 240, 365-371. Natl. Acad. Sci. USA 109, 3932-3937. Galliot, B. (2013). Injury-induced asymmetric cell death as a driving force for head Knowles, B. C., Weis, V. G., Yu, S., Roland, J. T., Williams, J. A., Alvarado, G. S., regeneration in Hydra. Dev. Genes Evol. 223, 39-52. Lapierre, L. A., Shub, M. D., Gao, N. and Goldenring, J. R. (2015). Rab11a Gao, N. and Kaestner, K. H. (2010). Cdx2 regulates endo-lysosomal function and regulates Syntaxin 3 localization and microvillus assembly in enterocytes. J. Cell epithelial cell polarity. Genes Dev. 24, 1295-1305. Sci. (in press) pii: jcs.163303. Gao, N., White, P. and Kaestner, K. H. (2009). Establishment of intestinal identity Kremer, J. R., Mastronarde, D. N. and McIntosh, J. R. (1996). Computer and epithelial-mesenchymal signaling by Cdx2. Dev. Cell 16, 588-599. visualization of three-dimensional image data using IMOD. J. Struct. Biol. 116, Gao, N., Davuluri, G., Gong, W., Seiler, C., Lorent, K., Furth, E. E., Kaestner, 71-76. K. H. and Pack, M. (2011). The nuclear pore complex protein Elys is required for Kuhnert, F., Davis, C. R., Wang, H.-T., Chu, P., Lee, M., Yuan, J., Nusse, R. and genome stability in mouse intestinal epithelial progenitor cells. Gastroenterology Kuo, C. J. (2004). Essential requirement for Wnt signaling in proliferation of adult 140, 1547-1555. small intestine and colon revealed by adenoviral expression of Dickkopf-1. Proc. Gasnereau, I., Herr, P., Chia, P. Z. C., Basler, K. and Gleeson, P. A. (2011). Natl. Acad. Sci. USA 101, 266-271. Identification of an endocytosis motif in an intracellular loop of Wntless protein, Lim, X., Tan, S. H., Koh, W. L. C., Chau, R. M. W., Yan, K. S., Kuo, C. J., van essential for its recycling and the control of Wnt protein signaling. J. Biol. Chem. Amerongen, R., Klein, A. M. and Nusse, R. (2013). Interfollicular epidermal stem 286, 43324-43333. cells self-renew via autocrine Wnt signaling. Science 342, 1226-1230. Gerges, N. Z., Backos, D. S. and Esteban, J. A. (2004). Local control of AMPA Lustig, B., Jerchow, B., Sachs, M., Weiler, S., Pietsch, T., Karsten, U., van de receptor trafficking at the postsynaptic terminal by a small GTPase of the Rab Wetering, M., Clevers, H., Schlag, P. M., Birchmeier, W. et al. (2002). Negative family. J. Biol. Chem. 279, 43870-43878. feedback loop of Wnt signaling through upregulation of conductin/axin2 in Goel, S., Chin, E. N., Fakhraldeen, S. A., Berry, S. M., Beebe, D. J. and colorectal and liver tumors. Mol. Cell. Biol. 22, 1184-1193. Alexander, C. M. (2012). Both LRP5 and LRP6 receptors are required to respond Ma, Y., Erkner, A., Gong, R., Yao, S., Taipale, J., Basler, K. and Beachy, P. A. to physiological Wnt ligands in mammary epithelial cells and fibroblasts. J. Biol. (2002). Hedgehog-mediated patterning of the mammalian embryo requires Chem. 287, 16454-16466. transporter-like function of dispatched. Cell 111, 63-75. Goldenring, J. R. (2013). A central role for vesicle trafficking in epithelial neoplasia: MacDonald, B. T. and He, X. (2012). Frizzled and LRP5/6 receptors for Wnt/β- intracellular highways to carcinogenesis. Nat. Rev. Cancer 13, 813-820. catenin signaling. Cold Spring Harb. Perspect. Biol. 4, a007880. Gregorieff, A. and Clevers, H. (2005). Wnt signaling in the intestinal epithelium: MacDonald, B. T., Tamai, K. and He, X. (2009). Wnt/beta-catenin signaling: from endoderm to cancer. Genes Dev. 19, 877-890. components, mechanisms, and diseases. Dev. Cell 17, 9-26. Gregorieff, A., Pinto, D., Begthel, H., Destree, O., Kielman, M. and Clevers, H. Markova, S. V., Golz, S., Frank, L. A., Kalthof, B. and Vysotski, E. S. (2004). (2005). Expression pattern of Wnt signaling components in the adult intestine. Cloning and expression of cDNA for a luciferase from the marine copepod Metridia Gastroenterology 129, 626-638. longa: A novel secreted bioluminescent reporter enzyme. J. Biol. Chem. 279, Habas, R., Kato, Y. and He, X. (2001). Wnt/Frizzled activation of Rho regulates 3212-3217. vertebrate gastrulation and requires a novel Formin homology protein Daam1. Cell Mastronarde, D. N. (2005). Automated electron microscope tomography using 107, 843-854. robust prediction of specimen movements. J. Struct. Biol. 152, 36-51. Habas, R., Dawid, I. B. and He, X. (2003). Coactivation of Rac and Rho by Wnt/ Meijering, E., Dzyubachyk, O. and Smal, I. (2012). Methods for cell and particle Frizzled signaling is required for vertebrate gastrulation. Genes Dev. 17, 295-309. tracking. Methods Enzymol. 504, 183-200. Haegebarth, A. and Clevers, H. (2009). Wnt signaling, Lgr5, and stem cells in the Mikels, A. J. and Nusse, R. (2006). Purified Wnt5a protein activates or inhibits beta- intestine and skin. Am. J. Pathol. 174, 715-721. catenin-TCF signaling depending on receptor context. PLoS Biol. 4, e115. Harterink, M., Port, F., Lorenowicz, M. J., McGough, I. J., Silhankova, M., Betist, Miyoshi, H., Ajima, R., Luo, C. T., Yamaguchi, T. P. and Stappenbeck, T. S. M. C., van Weering, J. R. T., van Heesbeen, R. G. H. P., Middelkoop, T. C., (2012). Wnt5a potentiates TGF-beta signaling to promote colonic crypt Basler, K. et al. (2011). A SNX3-dependent retromer pathway mediates regeneration after tissue injury. Science 338, 108-113. retrograde transport of the Wnt sorting receptor Wntless and is required for Wnt Montgomery, R. K., Carlone, D. L., Richmond, C. A., Farilla, L., Kranendonk, secretion. Nat. Cell Biol. 13, 914-923. M. E. G., Henderson, D. E., Baffour-Awuah, N. Y., Ambruzs, D. M., Fogli, L. K., Hattula, K., Furuhjelm, J., Tikkanen, J., Tanhuanpaa, K., Laakkonen, P. and Algra, S. et al. (2011). Mouse telomerase reverse transcriptase (mTert) Peranen, J. (2006). Characterization of the Rab8-specific membrane traffic route expression marks slowly cycling intestinal stem cells. Proc. Natl. Acad. Sci. linked to protrusion formation. J. Cell Sci. 119, 4866-4877. USA 108, 179-184. He, X., Saint-Jeannet, J.-P., Wang, Y., Nathans, J., Dawid, I. and Varmus, H. Muñoz, J., Stange, D. E., Schepers, A. G., van de Wetering, M., Koo, B.-K., (1997). A member of the Frizzled protein family mediating axis induction by Wnt- Itzkovitz, S., Volckmann, R., Kung, K. S., Koster, J., Radulescu, S. et al. 5A. Science 275, 1652-1654. (2012). The Lgr5 intestinal stem cell signature: robust expression of proposed He, X., Semenov, M., Tamai, K. and Zeng, X. (2004). LDL receptor-related proteins quiescent ‘+4’ cell markers. EMBO J. 31, 3079-3091. 5 and 6 in Wnt/beta-catenin signaling: arrows point the way. Development 131, Nusse, R., Fuerer, C., Ching, W., Harnish, K., Logan, C., Zeng, A., ten Berge, D. 1663-1677. and Kalani, Y. (2008). Wnt signaling and stem cell control. Cold Spring Harb. Henry, L. and Sheff, D. R. (2008). Rab8 regulates basolateral secretory, but not Symp. Quant. Biol. 73, 59-66. recycling, traffic at the recycling endosome. Mol. Biol. Cell 19, 2059-2068. Okamoto, M., Udagawa, N., Uehara, S., Maeda, K., Yamashita, T., Nakamichi, Y., Hong, W. and Tang, B. L. (1993). Protein trafficking along the exocytotic pathway. Kato, H., Saito, N., Minami, Y., Takahashi, N. et al. (2014). Noncanonical Wnt5a Bioessays 15, 231-238. enhances Wnt/beta-catenin signaling during osteoblastogenesis. Sci. Rep. 4, Huang, H. and He, X. (2008). Wnt/beta-catenin signaling: new (and old) players and 4493. new insights. Curr. Opin. Cell Biol. 20, 119-125. Pan, C.-L., Baum, P. D., Gu, M., Jorgensen, E. M., Clark, S. G. and Garriga, G. Huber, L. A., de Hoop, M. J., Dupree, P., Zerial, M., Simons, K. and Dotti, C. (2008). C. elegans AP-2 and retromer control Wnt signaling by regulating (1993a). Protein transport to the dendritic plasma membrane of cultured neurons mig-14/Wntless. Dev. Cell 14, 132-139. is regulated by rab8p. J. Cell Biol. 123, 47-55. Pinto, D., Gregorieff, A., Begthel, H. and Clevers, H. (2003). Canonical Wnt Huber, L. A., Pimplikar, S., Parton, R. G., Virta, H., Zerial, M. and Simons, K. signals are essential for homeostasis of the intestinal epithelium. Genes Dev. 17, (1993b). Rab8, a small GTPase involved in vesicular traffic between the TGN and 1709-1713. the basolateral plasma membrane. J. Cell Biol. 123, 35-45. Polakis, P. (2007). The many ways of Wnt in cancer. Curr. Opin. Genet. Dev. 17, Hume, A. N., Collinson, L. M., Rapak, A., Gomes, A. Q., Hopkins, C. R. and 45-51. Seabra, M. C. (2001). Rab27a regulates the peripheral distribution of Port, F. and Basler, K. (2010). Wnt trafficking: new insights into Wnt maturation, melanosomes in melanocytes. J. Cell Biol. 152, 795-808. secretion and spreading. Traffic 11, 1265-1271. Hutagalung, A. H. and Novick, P. J. (2011). Role of Rab GTPases in membrane Port, F., Kuster, M., Herr, P., Furger, E., Bänziger, C., Hausmann, G. and Basler, traffic and cell physiology. Physiol. Rev. 91, 119-149. K. (2008). Wingless secretion promotes and requires retromer-dependent cycling Jiang, M., Ku, W.-Y., Fu, J., Offermanns, S., Hsu, W. and Que, J. (2013). Gpr177 of Wntless. Nat. Cell Biol. 10, 178-185. regulates pulmonary vasculature development. Development 140, 3589-3594. Powell, A. E., Wang, Y., Li, Y., Poulin, E. J., Means, A. L., Washington, M. K., Kabiri, Z., Greicius, G., Madan, B., Biechele, S., Zhong, Z., Zaribafzadeh, H., Higginbotham, J. N., Juchheim, A., Prasad, N., Levy, S. E. et al. (2012). The Edison, Aliyev, J., Wu, Y., Bunte, R. et al. (2014). Stroma provides an intestinal pan-ErbB negative regulator Lrig1 is an intestinal stem cell marker that functions stem cell niche in the absence of epithelial Wnts. Development 141, 2206-2215. as a tumor suppressor. Cell 149, 146-158. Khandelwal, P., Prakasam, H. S., Clayton, D. R., Ruiz, W. G., Gallo, L. I., van Reya, T. and Clevers, H. (2005). Wnt signalling in stem cells and cancer. Nature Roekel, D., Lukianov, S., Peranen, J., Goldenring, J. R. and Apodaca, G. 434, 843-850. (2013). A Rab11a-Rab8a-Myo5B network promotes stretch-regulated exocytosis Rodriguez, O. C. and Cheney, R. E. (2002). Human myosin-Vc is a novel class V

in bladder umbrella cells. Mol. Biol. Cell 24, 1007-1019. myosin expressed in epithelial cells. J. Cell Sci. 115, 991-1004. DEVELOPMENT

2161 RESEARCH ARTICLE Development (2015) 142, 2147-2162 doi:10.1242/dev.121046

Roelink, H. and Nusse, R. (1991). Expression of two members of the Wnt family Tanaka, K., Kitagawa, Y. and Kadowaki, T. (2002). Drosophila segment polarity during mouse development–restricted temporal and spatial patterns in the gene product porcupine stimulates the posttranslational N-glycosylation of developing neural tube. Genes Dev. 5, 381-388. wingless in the endoplasmic reticulum. J. Biol. Chem. 277, 12816-12823. Rojas, R., van Vlijmen, T., Mardones, G. A., Prabhu, Y., Rojas, A. L., Tetteh, P. W., Farin, H. F. and Clevers, H. (2014). Plasticity within stem cell Mohammed, S., Heck, A. J. R., Raposo, G., van der Sluijs, P. and hierarchies in mammalian epithelia. Trends Cell Biol. 25, 100-108. Bonifacino, J. S. (2008). Regulation of retromer recruitment to endosomes by Tian, H., Biehs, B., Warming, S., Leong, K. G., Rangell, L., Klein, O. D. and de sequential action of Rab5 and Rab7. J. Cell Biol. 183, 513-526. Sauvage, F. J. (2011). A reserve stem cell population in small intestine renders Roland, J. T., Lapierre, L. A. and Goldenring, J. R. (2009). Alternative splicing in Lgr5-positive cells dispensable. Nature 478, 255-259. class V myosins determines association with Rab10. J. Biol. Chem. 284, van de Wetering, M., Sancho, E., Verweij, C., de Lau, W., Oving, I., Hurlstone, A., 1213-1223. van der Horn, K., Batlle, E., Coudreuse, D., Haramis, A.-P. et al. (2002). The Roland, J. T., Bryant, D. M., Datta, A., Itzen, A., Mostov, K. E. and Goldenring, beta-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal J. R. (2011). Rab GTPase-Myo5B complexes control membrane recycling and cancer cells. Cell 111, 241-250. epithelial polarization. Proc. Natl. Acad. Sci. USA 108, 2789–2794. van Es, J. H., Jay, P., Gregorieff, A., van Gijn, M. E., Jonkheer, S., Hatzis, P., Roth, S., Franken, P., Sacchetti, A., Kremer, A., Anderson, K., Sansom, O. and Thiele, A., van den Born, M., Begthel, H., Brabletz, T. et al. (2005). Wnt Fodde, R. (2012). Paneth cells in intestinal homeostasis and tissue injury. PLoS signalling induces maturation of Paneth cells in intestinal crypts. Nat. Cell Biol. 7, ONE 7, e38965. 381-386. Sakamori, R., Das, S., Yu, S., Feng, S., Stypulkowski, E., Guan, Y., Douard, V., van Es, J. H., Sato, T., van de Wetering, M., Lyubimova, A., Yee Nee, A. N., Tang, W., Ferraris, R. P., Harada, A. et al. (2012). Cdc42 and Rab8a are critical Gregorieff, A., Sasaki, N., Zeinstra, L., van den Born, M., Korving, J. et al. for intestinal stem cell division, survival, and differentiation in mice. J. Clin. Invest. (2012). Dll1(+) secretory progenitor cells revert to stem cells upon crypt damage. 122, 1052-1065. Nat. Cell Biol. 14, 1099-1104. Sakamori, R., Yu, S., Zhang, X., Hoffman, A., Sun, J., Das, S., Vedula, P., Li, G., Wallingford, J. B. and Habas, R. (2005). The developmental biology of Fu, J., Walker, F. et al. (2014). CDC42 inhibition suppresses progression of Dishevelled: an enigmatic protein governing cell fate and cell polarity. incipient intestinal tumors. Cancer Res. 74, 5480-5492. Development 132, 4421-4436. San Roman, A. K., Jayewickreme, C. D., Murtaugh, L. C. and Shivdasani, R. A. Wang, G. and Wu, G. (2012). Small GTPase regulation of GPCR anterograde (2014). Wnt secretion from epithelial cells and subepithelial myofibroblasts is not trafficking. Trends Pharmacol. Sci. 33, 28-34. required in the mouse intestinal stem cell niche in vivo. Stem Cell Rep. 2, 127-134. Wehrli, M., Dougan, S. T., Caldwell, K., O’Keefe, L., Schwartz, S., Vaizel- Sangiorgi, E. and Capecchi, M. R. (2008). Bmi1 is expressed in vivo in intestinal Ohayon, D., Schejter, E., Tomlinson, A. and DiNardo, S. (2000). arrow encodes stem cells. Nat. Genet. 40, 915-920. an LDL-receptor-related protein essential for Wingless signalling. Nature 407, Sato, N., Meijer, L., Skaltsounis, L., Greengard, P. and Brivanlou, A. H. (2004). 527-530. Maintenance of pluripotency in human and mouse embryonic stem cells through Willert, K. and Nusse, R. (2012). Wnt proteins. Cold Spring Harb. Perspect. Biol. 4, activation of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nat. a007864. Med. 10, 55-63. Willert, K., Brown, J. D., Danenberg, E., Duncan, A. W., Weissman, I. L., Reya, T., Sato, T., Mushiake, S., Kato, Y., Sato, K., Sato, M., Takeda, N., Ozono, K., Miki, Yates, J. R. and Nusse, R. (2003). Wnt proteins are lipid-modified and can act as K., Kubo, Y., Tsuji, A. et al. (2007). The Rab8 GTPase regulates apical protein stem cell growth factors. Nature 423, 448-452. localization in intestinal cells. Nature 448, 366-369. Wong, H.-C., Bourdelas, A., Krauss, A., Lee, H.-J., Shao, Y., Wu, D., Mlodzik, M., Sato, T., Vries, R. G., Snippert, H. J., van de Wetering, M., Barker, N., Stange, Shi, D.-L. and Zheng, J. (2003). Direct binding of the PDZ domain of Dishevelled D. E., van Es, J. H., Abo, A., Kujala, P., Peters, P. J. et al. (2009). Single Lgr5 to a conserved internal sequence in the C-terminal region of Frizzled. Mol. Cell 12, stem cells build crypt-villus structures in vitro without a mesenchymal niche. 1251-1260. Nature 459, 262-265. Wong, V. W. Y., Stange, D. E., Page, M. E., Buczacki, S., Wabik, A., Itami, S., van Sato, T., van Es, J. H., Snippert, H. J., Stange, D. E., Vries, R. G., van den Born, de Wetering, M., Poulsom, R., Wright, N. A., Trotter, M. W. B. et al. (2012). Lrig1 M., Barker, N., Shroyer, N. F., van de Wetering, M. and Clevers, H. (2011). controls intestinal stem-cell homeostasis by negative regulation of ErbB Paneth cells constitute the niche for Lgr5 stem cells in intestinal crypts. Nature 469, 415-418. signalling. Nat. Cell Biol. 14, 401-408. Sato, T., Iwano, T., Kunii, M., Matsuda, S., Mizuguchi, R., Jung, Y., Hagiwara, H., Wu, C.-h. and Nusse, R. (2002). Ligand receptor interactions in the Wnt signaling Yoshihara, Y., Yuzaki, M., Harada, R. et al. (2014). Rab8a and Rab8b are pathway in Drosophila. J. Biol. Chem. 277, 41762-41769. essential for several apical transport pathways but insufficient for ciliogenesis. Wu, X., Bowers, B., Rao, K., Wei, Q. and Hammer, J. A.III (1998). Visualization of J. Cell Sci. 127, 422-431. melanosome dynamics within wild-type and dilute melanocytes suggests a Schulte, G. (2010). International union of basic and clinical pharmacology. LXXX. paradigm for myosin V function In vivo. J. Cell Biol. 143, 1899-1918. The class Frizzled receptors. Pharmacol. Rev. 62, 632-667. Yan, K. S., Chia, L. A., Li, X., Ootani, A., Su, J., Lee, J. Y., Su, N., Luo, Y., Schulte, G. and Bryja, V. (2007). The Frizzled family of unconventional G-protein- Heilshorn, S. C., Amieva, M. R. et al. (2012). The intestinal stem cell markers coupled receptors. Trends Pharmacol. Sci. 28, 518-525. Bmi1 and Lgr5 identify two functionally distinct populations. Proc. Natl. Acad. Sci. Seabra, M. C., Mules, E. H. and Hume, A. N. (2002). Rab GTPases, intracellular USA 109, 466-471. traffic and disease. Trends Mol. Med. 8, 23-30. Yang, P.-T., Lorenowicz, M. J., Silhankova, M., Coudreuse, D. Y. M., Betist, M. C. Stefater, J. A., III, Lewkowich, I., Rao, S., Mariggi, G., Carpenter, A. C., Burr, and Korswagen, H. C. (2008). Wnt signaling requires retromer-dependent A. R., Fan, J., Ajima, R., Molkentin, J. D., Williams, B. O. et al. (2011). recycling of MIG-14/Wntless in Wnt-producing cells. Dev. Cell 14, 140-147. Regulation of angiogenesis by a non-canonical Wnt-Flt1 pathway in myeloid cells. Yin, X., Farin, H. F., van Es, J. H., Clevers, H., Langer, R. and Karp, J. M. (2014). Nature 474, 511-515. Niche-independent high-purity cultures of Lgr5(+) intestinal stem cells and their Strutt, D. I., Weber, U. and Mlodzik, M. (1997). The role of RhoA in tissue polarity progeny. Nat. Methods 11, 106-112. and Frizzled signalling. Nature 387, 292-295. Yu, H.-M. I., Jin, Y., Fu, J. and Hsu, W. (2010). Expression of Gpr177, a Wnt Sun, Y., Chiu, T. T., Foley, K. P., Bilan, P. J. and Klip, A. (2014). Myosin Va trafficking regulator, in mouse embryogenesis. Dev. Dyn. 239, 2102-2109. mediates Rab8A-regulated GLUT4 vesicle exocytosis in insulin-stimulated Yu, J., Chia, J., Canning, C. A., Jones, C. M., Bard, F. A. and Virshup, D. M. muscle cells. Mol. Biol. Cell 25, 1159-1170. (2014a). WLS retrograde transport to the endoplasmic reticulum during Wnt Takada, R., Satomi, Y., Kurata, T., Ueno, N., Norioka, S., Kondoh, H., Takao, T. secretion. Dev. Cell 29, 277-291. and Takada, S. (2006). Monounsaturated fatty acid modification of Wnt protein: its Yu, S., Nie, Y., Knowles, B., Sakamori, R., Stypulkowski, E., Patel, C., Das, S., role in Wnt secretion. Dev. Cell 11, 791-801. Douard, V., Ferraris, R. P., Bonder, E. M. et al. (2014b). TLR sorting by Rab11 Takeda, N., Jain, R., LeBoeuf, M. R., Wang, Q., Lu, M. M. and Epstein, J. A. endosomes maintains intestinal epithelial-microbial homeostasis. EMBO J. 33, (2011). Interconversion between intestinal stem cell populations in distinct niches. 1882-1895. Science 334, 1420-1424. Yu, S., Yehia, G., Wang, J., Stypulkowski, E., Sakamori, R., Jiang, P., Tamai, K., Semenov, M., Kato, Y., Spokony, R., Liu, C., Katsuyama, Y., Hess, F., Hernandez-Enriquez, B., Tran, T. S., Bonder, E. M., Guo, W. et al. (2014c). Saint-Jeannet, J.-P. and He, X. (2000). LDL-receptor-related proteins in Wnt Global ablation of the mouse rab11a gene impairs early embryogenesis and signal transduction. Nature 407, 530-535. matrix metalloproteinase secretion. J. Biol. Chem. 289, 32030-32043. DEVELOPMENT

2162